A Novel Quasi-3D Landslide Dynamics Model: From Theory to Applications and Risk Assessment

Abstract

Abstract Submarine landslides can pose a significant threat to offshore installations and coastal communities. They can strike installations far from their origin and generate destructive tsunamis. To assess and quantify this hazard, it is necessary to be able to model their dynamics in complex submarine environments with realistic rheological input parameters. Many submarine landslides involve cohesive visco-plastic soils, which can be described mathematically by rheological models such as the non-linear Herschel–Bulkley model. To model these events, accounting for complex bathymetry and rheological behavior, NGI has developed BingClaw. It incorporates buoyancy, hydrodynamic resistance and remolding, which are crucial for underwater landslide dynamics. BingClaw has been used to study the dynamics and tsunami generation of some of the largest and most complex submarine landslides in the world such as the Storegga Slide about 8000 years ago and the 1929 Grand Banks landslide and tsunami. In both cases, BingClaw provided a far more realistic description of both the landslide dynamics and the induced tsunami than other models. The link to the tsunami generation was used to better constrain the landslide dynamics. Here, we demonstrate how BingClaw is used for geohazard applications, including attempts to hindcast past landslides directly relevant to these applications, as often rheological data are sparse or not available from a given site. We first present benchmark results comparing the landslide model with results from laboratory experiments. Then, we show comparisons between simulations and observed landslide run-out for both offshore and onshore applications. The onshore application provides additional well-controlled field studies for validation, in soils with high sensitivity. We also used BingClaw in an offshore/nearshore geohazard project, namely the Bjørnafjorden project offshore western Norway. There, we linked the run-out analysis directly to static and seismic slope stability evaluations, and the predicted run-out scenarios were used in the assessment of competing bridge concepts and their foundations in the deep fjord (around 560 m water depth). These studies illustrate that this novel method, applicable for onshore and offshore geohazard assessments, can reliably reproduce field observations using realistic rheological parameters, which is important when estimating the risk posed by submarine mass movements, particularly with respect to the potential impact on infrastructure.